Using Deep and Convolutional Neural Networks for Accurate Emotion Classification on DEAP Dataset

By S. Tripathi, Shrinivasa G Acharya, Ranti Dev Sharma, Sudhanshu Mittal, S. Bhattacharya. AAAI Conference on Artificial Intelligence, 2017

350 citations recorded by the source corpus.

Emotion recognition is an important field of research in Brain Computer Interactions. As technology and the understanding of emotions are advancing, there are growing opportunities for automatic emotion recognition systems. Neural networks are a family of statistical learning models inspired by biological neural networks and are used to estimate functions that can depend on a large number of inputs that are generally unknown. In this paper we seek to use this effectiveness of Neural Networks to classify user emotions using EEG signals from the DEAP (Koelstra et al (2012)) dataset which represents the benchmark for Emotion classification research. We explore 2 different Neural Models, a simple Deep Neural Network and a Convolutional Neural Network for classification. Our model provides the state-of-the-art classification accuracy, obtaining 4.51 and 4.96 percentage point improvements over (Rozgic et al (2013)) classification of Valence and Arousal into 2 classes (High and Low) and 13.39 and 6.58 percentage point improvements over (Chung and Yoon(2012)) classification of Valence and Arousal into 3 classes (High, Normal and Low). Moreover our research is a testament that Neural Networks could be robust classifiers for brain signals, even outperforming traditional learning techniques.

Subfields: Artificial Intelligence, Cognitive Neuroscience

View Paper: https://www.semanticscholar.org/paper/0be8156073a1a586badb88a93f453915ca5e927a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Shuai Wang
2. B. Liu
3. Lei Zhang
4. Bing Liu
5. T. Mcginnity
6. Bin Hu
7. Junjie Zhu
8. Amir Hussain

In Cognitive Neuroscience

1. T. Mcginnity
2. Bin Hu
3. Amir Hussain
4. Mufti Mahmud
5. M. S. Kaiser
6. Prayag Tiwari
7. Yazhou Zhang
8. Meihong Yang

Citations in the corpus

• EEG Based Emotion Recognition: A Tutorial and Review (2022, methodology): “[132] extracted nine types of statistical EEG characteristics of signals as the input of the CNN model, and inally, the efect of this method reached and exceeded those of mainstream methods.”
• An Efficient LSTM Network for Emotion Recognition From Multichannel EEG Signals (2020, background): “Despite the popularity of subjectdependent models in EEG-based emotion recognition [19], [20], [21], some recent studies [22], [23], [24], [25], [26], [27], [28] suggest building models specially designed for subjectindependent evaluation.”
• Deep Learning in Mining Biological Data (2020, methodology): “eart LV segmentation Signals MAHNOB-HCI DA [81] Motion action decoding BCI Competition IV DBN [82], CNN [83{85] Motion action decoding DEAP dataset DBN [86,87] Motion action decoding DEAP dataset CNN [88] Emotion classification Freiburg dataset CNN [89] Seizure prediction Ninapro database DBN [90], ”
• Physiological Signals-based Emotion Recognition via High-order Correlation Learning (2019, methodology): “As Table 1 shows, the raw-data construct is as follows: In the video/trial * channel *reading, we employ different methods based on psychophysiological findings to extract different modalities of physiological signals from the DEAP dataset, following the settings in References [37, 39].”
• Applications of Deep Learning and Reinforcement Learning to Biological Data (2017, methodology): “explored DNN (with Softmax activator and Dropout) and CNN [198] (with Tan Hyperbolic, Max Pooling, Dropout, and Softplus) for emotion classification from the DEAP dataset using EEG signals and response face video [175].”
• Deep learning for sentiment analysis: A survey (2018, methodology): “Tripathi et al.(135) used deep and CNN-based models for emotion classification on a multimodal dataset DEAP, which contains electroencephalogram and peripheral physiological and video signals.”